The long-term objective of our research is to understand visual signaling by the primate retina, and to exploit this knowledge in the treatment of blindness. The goal of the proposed work is to understand three fundamental aspects of visual signaling in the major ganglion cell types of the retina, which convey distinct visual signals to multiple targets in the brain. The specific aims are to explore (1) how the elementary signal for daylight vision, the activation of a single cone, propagates through parallel pathways to the distinct ganglion cell types; (2) the spatial structure and origin of nonlinear visual computations performed in the inner retinal circuitry; (3) the organization and function of polyaxonal amacrine cells which modulate visual signals in ganglion cells. To approach these problems, we will exploit our unique large-scale multi- electrode recordings from isolated primate retina to sample complete visual signals in the major 5 ganglion cell types (ON parasol, OFF parasol, ON midget, OFF midget, and small bistratified), which collectively constitute 75% of the visual representation. We will also extend this unique experimental approach with three novel techniques: independent stimulation of each of the cone photoreceptors over a large region of retina, advanced statistical analysis to uncover the spatial structure of nonlinear computations at the resolution of individual cones, and electrical imaging to record from several types of amacrine cells simultaneously with ganglion cells. At the completion of this work we hope to have a deeper understanding of how cone signals, transmitted through the parallel pathways, are combined, processed and modulated in the parallel circuits of the primate retina.